Biomarkers of outcome in paediatric pulmonary hypertension

In the era of goal-driven therapies, the ability of predicting prognosis is essential in the management of patients with pulmonary hypertension (PH). Since paediatric PH differs from the adult form, what applies to adults does not necessarily translate to children as well and, unfortunately, there are no sufficient studies conducted in children, neither for prognostic factors nor for treatment. In PH, cardiac magnetic resonance (CMR) is the most accurate way of assessing RV function and pulmonary haemodynamics non-invasively. Given that prognosis is heavily influenced by RV function, CMR may also prove useful in providing prognostic information. While the prognostic role of CMR is well-established in adults, to date there is only one study demonstrating this correlation in children. The aim of this study was to look if certain CMR-derived variables are able to predict outcomes in paediatric PH. Learning which biomarkers are useful in estimating outcomes in children could be the basis for future improvements in the management of these patients. The study was conducted at Great Ormond Street Hospital in London. The study population included 226 patients diagnosed with PH with a median age of 8.79 years. All images were acquired using real-time radial k-t SENSE CMR, which allows to scan children in shorter times and without breath holding, thus avoiding the need for general anaesthesia. For each patient, measures of RV and LV function were assessed, and mPAP was estimated non-invasively thanks to its correlations with septal curvature. During follow-up, 26 patients died and 16 underwent lung transplantation. Among all CMR parameters, those that correlated the most with survival on univariate analysis were RVEF, mPAP, SCR, RVMi and HR. This was also reflected in tertile-based Kaplan-Meier survival curves for these variables: being in the lowest tertile for SCR and RVEF and in the highest tertile of mPAP, RVMi and HR was correlated to a poorer prognosis. With a conditional inference tree, it was possible to determine that the only independent biomarkers of outcome are RVEF and mPAP, and it was also possible to define specific cut-off for a worse outcome (RVEF ≤ 39% and mPAP > 56.124 mmHg). In conclusion, prognosis in paediatric pulmonary hypertension is heavily influenced by RVEF and mPAP, both of which can be accurately assessed with CMR. CMR is a useful diagnostic and prognostic tool in children with PH, and it could also be a useful adjunct in the management of these patients.